Abstract

Daily efavirenz 400 mg (EFV400) was virologically noninferior to 600 mg (EFV600) at 48 weeks in treatment-naive patients. We evaluated EFV400 and EFV600 pharmacokinetics (NONMEM v. 7.2), assessing patient demographics and genetic polymorphisms (CYP2B6, CYP2A6, CYP3A4, NR1I3) as covariates and explored relationships with efficacy (plasma HIV-RNA (pVL) 200 copies/mL) and safety outcomes at 48 weeks in 606 randomized ENCORE1 patients (female = 32%, African = 37%, Asian = 33%; EFV400 = 311, EFV600 = 295). CYP2B6 516G>T/983T>C/CYP2A6*9B/*17 and weight were associated with efavirenz CL/F. Exposure was significantly lower for EFV400 (geometric mean ratio, GMR; 90% confidence interval, CI: 0.73 (0.68-0.78)) but 97% (EFV400) and 98% (EFV600) of evaluable pVL was 200 copies/mL at 48 weeks (P = 0.802). Four of 20 patients with mid-dose concentrations 1.0 mg/L had pVL >= 200 copies/mL (EFV400 = 1; EFV600 = 3). Efavirenz exposure was similar between those with and without efavirenz-related side effects (GMR; 90% CI: 0.95 (0.88-1.02)). HIV suppression was comparable between doses despite significantly lower EFV400 exposure. Comprehensive evaluation of efavirenz pharmacokinetics/pharmacodynamics revealed important limitations in the accepted threshold concentration.